EP1317090A1 - Method and apparatus to code different data packets for different receivers - Google Patents

Abstract

The method involves sending data or data streams (D1,2) from a transmitting station (BS) via an interface to receiving stations (MS1-MS3) after encoding with redundancy. The data for at least two receiving stations are combined before encoding. Code data from the combined data or from the coded data produced from it are produced with unequal redundant data parts for the different receiving stations. <??>AN Independent claim is also included for the following: a transmitter and a receiver for implementing an inventive method.

Description

The invention relates to a method for coding Package data with the generic features of the claim 1 or a device for performing such Process.

In the case of mobile radio systems, for example in accordance with the GSM standard (Global System for Mobile communication) or UMTS (Universal Mobile Telecommunications System) data or Data packets from a network-side station to subscriber-side Stations are sent before sending pro Participants or receivers channel-coded against channel interference and transmitted separately for each participant or recipient. The one there The error rate achieved depends on the channel and the used one Code. Improve the error correction properties for channel coding with the length of the channel code. On Code with a large block length can be inserted of more redundancy correspondingly a better error correction property as a short code.

Channel encodings are both downward (Downlink), i.e. from a network-side station to one subscriber-side station, as well as in the upward direction (Uplink), i.e. in the direction of a station on the subscriber side to a network-side station, each separately for each connection on the data to be transmitted to protect against channel interference applied. It is also known from the UMTS in Downward direction to summarize the data for several participants and code together. This will make the corresponding one Overall, the code is correspondingly longer and consequently the Bit error rate per data for a particular receiver is lower.

The object of the invention is such a method to improve data coding or a device to propose to carry out such a procedure.

This task is accomplished by a method of encoding data with the features of claim 1 or devices solved with the features of claims 5 and 6, respectively. advantageous Refinements are the subject of dependent claims.

Such a method of providing and transferring redundantly coded data in a communication system, at the data coded for transmission via the interface the previously summarized data or from data generated from it encoded data with differently sized redundant data components generated for the various receiver-side stations accordingly includes the possibility of a Weighting the redundancy for individual receiver stations or some of the data to be sent before, during or after the actual coding to add redundancy the data.

To generate a larger proportion of redundant data in addition to a repeat procedure, any other suitable method, e.g. another, additional channel coding for the corresponding data components or already coded data components.

Fig. 1

schematisch eine Anordnung von miteinander kommunizierenden Stationen und zwischen diesen ausgetauschten Datenpaketen;

Fig. 2

zwei Blockdarstellungen zum Veranschaulichen eines für sich bekannten Codierungsverfahrens;

Fig. 3

ein Blockschema zum Veranschaulichen eines vorteilhaften Codierungsverfahrens.

An embodiment is explained below with reference to the drawing. Show it:

Fig. 1

schematically shows an arrangement of communicating stations and data packets exchanged between them;

Fig. 2

two block diagrams to illustrate a coding method known per se;

Fig. 3

a block diagram to illustrate an advantageous coding method.

As can be seen from Fig. 2a, conventional coding methods for performing channel coding data D1 or data packets that a first receiver or a first are assigned to the receiving station MS1, thereby encoding a data packet C1 to be sent arises, which consists of Data D1 and redundant data R1 exist. Likewise, with Move data D2 for a second receiver MS2.

In an improved embodiment, as can be seen from FIG. 2b, Data D1 for a first subscriber or recipient MS1 and data D2 for a second subscriber or receiver MS2 arranged one behind the other and coded together. The resulting data sequence for sending via the Interface consists of data D1 for a first receiver, Data D2 for a second receiver and another Data block with redundants common to both receivers Data R3. In principle, this scheme is also more than that Data expandable for two recipients.

As can be seen from FIG. 1, such an application is used Method for coding to different recipients or receiving stations MS1 - MS3 data to be sent D1 - D3 in particular in the downward direction. In a network side, sending station BS are data or data packets D1, D2, D3 are provided for transmission via a radio interface V. This data D1 - D3 can be in the network-side station BS generated or from other facilities to this network side Station BS have been transferred.

For the temporary storage of data or data packets D1-D3 has the network-side station BS, which follows is also referred to as the sending station, a memory or Memory section M-BS. Furthermore, the sending Station BS a control device CT and a transmission device or transceiver T for processing, control and performing the transmission operation via the radio interface V on.

The data D1-D3 provided in the sending station BS are coded together, i.e. provided with redundancy and as redundantly coded data C1 '- C3' (1st in Fig. 1) the radio interface V to the receiver side or receiving Stations MS1, MS2, MS3 transmitted.

These stations MS1-MS3 on the receiver side also point facilities necessary for their operation, in particular a control device CT, a receiving or transmitting / receiving device R and a memory or storage section M-MS for temporary storage of received encoded data packets or data C '= C1' + C2 '+ C3'. All receiving stations MS1 - MS3 receive the whole Coded data sequence C '. This coded data sequence C 'is then in the individual receiving stations MS1 - MS3 each decoded and for the respective station MS1, MS2 or MS3 are those determined for this station MS1, MS2 or MS3 Portions of the reconstructed data D1, D2 and D3 are extracted. This can optionally also be carried out directly during the decoding become.

By channel coding, these data packets are D1, D2 and D3 Redundancy R1, R2 and R3 added. This results in that Codeword C 'to be transmitted via the radio interface V or previously C = "C1 + C2 + C3" from the data to be transmitted D1 and their redundancy R1 (or D2 and R2, D3 and R3). Through the advantageous summary of data for different recipients the user data actually to be transmitted or Net data combined into a common data block. This composite data block D is then shared with the Data encoded so that a code word C with the data D = D1 + D2 + ... and a redundancy R = R1 + R2 + ... arises. In particular, the total redundancy R for all receiver-side stations Ms1, Ms2, ... for reconstruction the original data to be sent is used at the same time become.

As can be seen more clearly from FIG. 3, it is not only possible Data for various MS1-MS3 receiver stations summarize before coding, but also a different one Distribution for the different data streams, Data or data blocks D1 - D3 to carry out during coding. This allows the redundancy for the individual data streams to be corresponding the actual channel properties on each Connections on the radio interface V through different distribution can be adapted appropriately.

In the embodiment shown in Fig. 3, the Data D1 for the receiver-side stations MS1-MS3 in usual Way coded and transmitted together. In the Corresponding to the first line in FIG. 3 are the data D = D1 + D2 + D3 for the three participants or the three receiver side Stations MS1, MS2 and MS3 shown. In the second Coded data C = C1 + C2 + C3 are shown on the line, which result from the channel coding. The visual separation in the drawing is only for illustration, that data shares from different data sources or for different Recipients are present. In fact, the individual Percentage of data and / or encoded data according to the coding method, sometimes with each other mixed and / or coded.

However, coded ones are provided via the radio interface V Transfer data C 'with additional processing, as can be seen from the third line.

The encoded data C 'for the second subscriber station In the illustrated embodiment, MS2 are said to have good conditions can be transferred so few or no errors occur. Accordingly, the receiver side Correction in the second station on the receiver side MS2 requires little or no redundancy. As from the middle row can be seen from the Share the data D2 for the second receiver-side station MS2 encoded data C2 provided, but not in this Transfer scope. The coded data C2 for the second receiver side Station MS2 are used for example for known puncturing to a lower Data amount C2 'reduced and for transmission in the data stream C 'provided as can be seen from the bottom line is. Alternatively, it is also possible to use one instead of one usual coding and subsequent puncturing directly encoded data C2 'punctured to the original data D2 ready to provide. Even when displaying the third line from Fig. 3 serves the separation into individual coded data blocks C1 '- C3' again only to illustrate the principle.

The freely available block length for others to be transmitted encoded data has become larger. Can accordingly original data D3 for the third receiver-side station MS3 encoded in the usual way to data C3 and then by means of a kind of frame filling, which is also called repeating or bit-stuffing is provided become. In other words, the redundancy for the data D3 by providing a larger coded number of Data C3 'for transmission to the third party at the recipient end Station MS3 can be increased. Alternatively, of course, is also direct the generation of a correspondingly larger amount of coded data of data C3 'from the original data D3 possible.

In particular, the original is summarized beforehand Data D1 - D3 for the different receiver side Stations MS1 - MS3 before coding, such as this is explained with reference to Fig. 2b.

Advantageously, such a method can be used redundancy is reduced for a number of connections, to increase redundancy for other connections. This can result in more errors on these other connections corrected the originally assigned data blocks become. For example, one code for each separate data stream or data block correct three errors, the new code with a double code word length for the two corrected combined data flows a total of six errors. By redistribution it is also possible, in particular, to Number of correctable errors according to the actual ones or expected needs of the two data streams redistribute.

The use of so-called convolutional codes as they are taken on their own are known can be done in a comparable manner. The data D1, D2, D3, ... from two or more data streams for different receiver stations MS1, MS2, MS3, ... are temporarily stored in the memory M-BS on the receiver side, arranged one after the other and then with convolutionally coded the same code. After the data D1, D2, D3, ... convolutionally coded as coded data C = C1 + C2 + C3 + ... are provided for the data shares of the individual Data streams D1, D2, D3, ... no action as required, a puncturing or a repetition carried out. This is also called rate matching, i.e. Adaptation referred to the available data rate. For the decision, whether to perform puncturing or repetition the functions of the channel properties or the distances to the relevant receiver stations or receivers MS1 - MS3 are taken into account. Conveniently, when using puncturing or repetition the total of the coded data C = C1 + C2 + C3 left the same.

To illustrate this can be combined in one first case data for a first participant or a first receiver-side station MS1 with a binary 3-error-correcting BCH code (30,15,7) (BCH: Bose, Chaudhuri, Hocquenhem) be encoded. In this example, the block length would be n = 30 bits and the length of the information data k = 15 bits, so that the number of redundant Bit to n - k = 15 results. The so-called Hamming distance determines be twice the number of errors allowed plus 1, in this example seven. In this 15 bits of redundancy become 15 information bits added so that the code word to be transmitted is a Has a block length of n = 30. Below are the data blocks two participants combined into a data block of 30 bits and with a 6-error correcting binary BCH code, z. B. (63,30,13). This code then has one Correctability of six errors with an almost identical one Code rate, therefore, is compared to the application of two 3-error correcting BCH codes (30,15,7) advantageous.

The use of a so-called for itself is even cheaper known Reed-Solomon (RS) codes. Is used to encode the individual data for the various data connections RS code (31, 25, 7) is used, this results in comparison when summarizing the data for two connections resulting RS code of (62,50,13). It is advantageous it stops when using a common code important where, that is, on which of the connections to the individual receiver stations MS1, MS2, ..., Errors have occurred. The receiver-side stations MS1, MS2, ... must of course use a summarized Have coding according to knowledge or have informed be processed in accordance with the received To be able to make data.

Claims (6)

Method for providing and transmitting data in a communication system (GSM), in which Data or data streams (D1, D2) are transmitted from a transmitter-side station (BS) via an interface (V) to receiver-side stations (MS1, MS2, MS3) communicating with them, wherein for the data to be sent (D1, D2, D3), coded data (C1, C2, C3) with redundancy (R1, R2) are generated on the transmitter side before the transmission and the data (D1, D2, D3) for at least two of the receiver-side stations (MS1, MS2, MS3) being combined on the transmitter side before generating the coded data (C1, C2, C3) and from the combined data (D = D1 + D2 + D3) encoded data (C = C1 + C2 + C3) are generated with redundancy, characterized in that for sending via the interface (V) coded data (C '= C1' + C2 '+ C3') from the summarized data (D = D1 + D2 + D3) or from coded data generated from it (C = C1 + C2 + C3 ) with differently sized redundant data components (C2 'or C3') for the different receiver-side stations (MS2, MS3). The method of claim 1, wherein
a puncturing method is used to generate a smaller portion of redundant data (C2 '). The method of claim 1 or 2, wherein
a repetition method or a further channel coding is used to generate a larger proportion of redundant data (C3 '). A method according to any preceding claim, in which the total number of encoded data (C '= C1' + C2 '+ C3') when generating this coded data (C '= C1' + C2 '+ C3') with differently sized redundant data components (C2 'or C3') equal to the total number of shares otherwise Coded data generated unchanged (C = C1 + C2 + C3) with redundant data components of the same size (C2 or C3) becomes. Transmitting device (BS) for performing a method according to any preceding claim a memory (M-BS) for buffering data to be coded (D1 - D3) for different receiver stations (MS1, MS2 or MS3) and a control device (CT) for assembling the data to be coded (D1 - D3) for several receiver stations (MS1 MS3) and for generating redundantly coded data (C1 '- C3') from the composite data. Receiver device (BS) for performing a method according to one of claims 1-4 with a memory (M-BS) for buffering received, redundantly coded data (C = C1 '+ C2' + C3 ') and a control device (CT) for processing the received coded data (C = C1 '+ C2' + C3 ') for reconstructing error-free original data (D1, D2 or D3) from the composite and coded data (C = C1' + C2 ' + C3 ').

Images